|
Article Excerpt
ARTICLE CONTENTS
INTRODUCTION I. THE PATH OF LEGISLATION A. The VRARA in the House: Channeling Dissent into Failed Amendments B. The VRARA in the Senate: Channeling Dissent into Postenactment Legislative History II. THE EVIDENTIARY RECORD A. Rates of Minority Voter Turnout, Registration, and Officeholding B. Evidence Concerning the Preclearance Process: Rates of Submissions, Denials, and Requests for More Information C. Voting Rights Violations in Covered Jurisdictions III. WHY THE BASIC STRUCTURE OF THE LAW REMAINS INTACT A. Retaining the Same Coverage Formula B. Bailout C. The Section 5 Enforcement Regime IV. THE NEW STANDARD FOR RETROGRESSION A. Preferred Candidates of Choice 1. Minority Candidate 2. Democratic Candidate 3. The DOJ's Most Recent Approach and the One Likely To Be Followed 4. Candidates Uniquely Preferred by Minorities 5. The Role of Incumbency in Determining Candidate of Choice B. The Ability To Elect 1. No More Tradeoffs for Influence Districts 2. "Naturally Occurring" Majority-Minority Districts? 3. The Importance of Racial Bloc Voting to the Ability To Elect 4. "Ability To Elect" as a Continuous or Dichotomous Variable? C. Diminishing 1. Ability To Elect per District or Across Districts? 2. Diminution Through Overconcentration and Underconcentration of Minority Voters 3. The Art and Science of Measuring Diminution in the Ability To Elect CONCLUSION
INTRODUCTION
In the series of cases that have made up the Supreme Court's recent jurisprudence concerning congressional power to protect civil rights, the Voting Rights Act (VRA) has been the standard against which all other statutes are judged. (1) Unlike other civil rights statutes that may have swept too broadly in their geographic reach and permanent duration, section 5 of the VRA (2) targeted a limited number of geographically defined wrongdoers (3) and did so for a limited time. (4) Unlike those constitutionally precarious statutes that may have elevated classes from the lower tiers of Fourteenth Amendment scrutiny, (5) the VRA dealt specifically with race (a classification that is "immediately suspect" (6)) and with voting ("a fundamental political right, because [it is] preservative of all rights" (7)). And unlike those statutes with a tenuous relationship to unconstitutional state action, (8) the VRA was built on a record of persistent constitutional violations by state actors who were unapologetic in their defiance of federal court orders. (9)
Because Congress acted at the apex of its power to enforce the guarantees of the post-Civil War Amendments in passing the VRA, the Court could stomach the tough medicine that is section 5 of the Act. That measure stands alone in American history in its alteration of authority between the federal government and the states and the unique procedures it requires of states and localities that want to change their laws. No other statute applies only to a subset of the country and requires covered states and localities to get permission from the federal government before implementing a certain type of law. (10) Such a remedy was necessary because case-by-case adjudication of voting rights lawsuits proved incapable of reining in crafty Dixiecrat legislatures determined to deprive African Americans of their right to vote, regardless of what a federal court might order. (11)
Congress intended the expiration of section 5 to force the nation to take stock of its progress, or lack thereof, in achieving equal voting rights, as well as to adapt the law to new challenges and changing political realities. Those who originally crafted the law, however, could not have foreseen how section 5 would become, in both substance and symbolism, a cornerstone of the architecture of federal election law and civil rights guarantees. As each election reminds us of how far we need to go in securing the equal right to vote, the notion that we might allow this most successful of civil rights protections to die on the vine has become so unacceptable that Congress has now reauthorized this "emergency" provision for another twenty-five years.
Elsewhere I have described how the VRA could have been transformed to address the problems facing minority voters that constituted the principal justification for the law. (12) Even for many who favored renewal, the reauthorization process in the summer of 2006 represented a missed opportunity to deal with some of these problems. In addition to the political and judicial constraints placed on the reauthorization debate, the specter of returning to an age and political environment first disciplined by section 5 of the VRA paralyzed any attempt to use this opportunity to address the most pressing voting rights challenges.
This Article attempts to explain the constraints on the process that led the law to take the form that it did and to identify the best evidence in the legislative record to ensure the law's constitutionality. More important from the standpoint of those wishing to interpret or enforce this new law, this Article provides an interpretation of the law's key provision that would allow it to do more good than harm. Part I provides a summary of the unique legislative history of the Fannie Lou Hamer, Rosa Parks, and Coretta Scott King Voting Rights Act Reauthorization and Amendments Act of 2006 (the VRARA, or the new section 5). (13) As with other legislation, disagreements about the statute's meaning were passed on to the courts, and various legislators attempted to manipulate legislative history for partisan ends. Never before in American history, however, has a Senate committee that unanimously voted in favor of a law later published a postenactment committee report that was supported only by members of one party. Part II examines the sufficiency of the evidentiary record assembled by Congress to justify the continued operation of section 5 in the areas that it currently covers (the "coverage formula" or "trigger"). Section 5 applies only to certain parts of the country, based on voting practices and data that are at least thirty years old. The novel constitutional question posed by this law was how Congress could provide a record of constitutional violations necessitating the continuation of a law that, if it works as intended, prevents such evidence from emerging. Part III explains why the law had to take the form that it did, despite widespread concerns about the coverage formula and statutory architecture. Various proposals to tinker with the well-known section 5 procedures--namely, where it should apply, how jurisdictions might escape its constraints, and how the law would be administered--faced an uphill battle and were soundly defeated. Part IV forms the bulk of this Article and offers an interpretation of the key provision of the new section 5: the new retrogression standard that prevents covered jurisdictions from enacting or administering voting laws that "diminish" minority voters' "ability ... to elect their preferred candidates of choice." (14) The proposed interpretation, which focuses on the extent of racial polarization in the electorate, represents an attempt to save the law from likely constitutional challenges and from enforcement patterns that would contradict the underlying purposes of the law.
The Conclusion presents an argument presaged here. The descriptive and normative sections of this paper are united by a common appreciation for the unprecedented political context in which this reauthorization occurred. In a sense, as compared to the legislative efforts of twenty-five and forty-two years ago, everything should have been different for this reauthorization. The constitutional test the courts would now apply never before cast a shadow on legislative bargaining. The partisan stakes, never before accorded much significance, were now well known. The most salient threats to minority voting rights had evolved beyond the categories and geography contemplated by the VRA. Nevertheless, the fear and uncertainty of what the world would be like without it allowed transformation only in the direction of restoring the Act to its original meaning. The consequences of this altered political reality do not end with the passage of the law, however. Despite historically familiar purposes and language, the VRA begins to mean something different when grafted onto a political system that it helped shape.
I. THE PATH OF LEGISLATION
When considered in the abstract, and against the history of American election reform and civil rights legislation, the 2006 reauthorization of the VRA by a near unanimous vote represents a remarkable, even if predictable, achievement. (15) The vote was remarkable in that almost all participants in the policy debate recognize that section 5 of the VRA represents a unique exception to the normal functioning of federalism and partisan politics; many Republicans consider it offensive to their notions of color-blindness and states' rights, and some Democrats see it as counter to their political interests. (16) The vote was predictable, however, in that virtually no one wanted to be on record opposing the legislation. (17) Republicans who may have disagreed with the legislation in principle nevertheless viewed it as largely serving their political interests. Most of them considered redistricting pursuant to aggressive enforcement of section 5 as creating inefficient Democratic districts. Moreover, the legislation appeared to be a relatively costless step toward thawing relationships with African Americans and maintaining gains among Hispanic supporters. On the other hand, most Democrats supported the reauthorization in principle, and those who did not considered opposition (or even amendment) to constitute political suicide.
A. The VRARA in the House: Channeling Dissent into Failed Amendments
Despite the widespread consensus in favor of reauthorization, many potential obstacles could have derailed the steady progress toward renewal. (18) One cannot overstate the importance of the unlikely leadership of James Sensenbrenner, Chairman of the House Committee on the Judiciary, in pushing through the legislation. (19) He wanted the reauthorization of the Voting Rights Act to occur on his watch, and, consistent with his leadership style on other issues, nothing was going to stand in his way. Pursuant to the rules of the Republican Conference, however, his term as Chairman was to expire at the end of 2006, a year before section 5 of the VRA was scheduled to sunset. (20) Therefore, the timetable for the legislation was moved up a year, with House hearings held between October 2005 and May 2006.
On the House side, as is often true with legislation expected to pass by a wide margin, the hearings featured relatively few witnesses testifying against reauthorization. (21) Most of the effort of the House Committee was directed at assembling a record demonstrating the persistence of discrimination in voting in the covered jurisdictions and the constitutionality of the proposed bill. Nevertheless, once the measure moved toward the floor, two sets of Republicans became more vocal concerning their objections to the bill. One group, mainly from the Southern jurisdictions covered by section 5, renewed familiar arguments from the VRA's inception that their states were unfairly targeted and urged reforms to the coverage formula and bailout procedures. (22) Members of another faction objected to the language assistance provisions of section 203 of the VRA, (23) echoing similar arguments expressed during the concurrent debate over immigration reform. (24) Although the original rule for the vote on the VRA would have precluded any amendments, those with objections invoked the norm, if not the rule, of the Republican Conference that a "majority of the majority" had to support a bill before it could be moved to the floor. (25) To mollify the holdouts, four amendments were allowed on the floor. These ranged from proposals to alter the coverage formula (26) or bailout procedures (27) to others attempting to accelerate the sunsetting of the law (28) or to change the language assistance provisions. (29) All were soundly defeated, but three garnered a majority of the Republican members. (30) The House passed the bill by a vote of 390 to 33 on July 13, 2006. (31)
B. The VRARA in the Senate: Channeling Dissent into Postenactment Legislative History
The story in the Senate was quite different. The nine Senate Judiciary Committee hearings held between April 27, 2006, and July 13, 2006, featured heated debates concerning the constitutionality and desirability of the legislation. (32) Individual Judiciary Committee members had serious reservations about the proposed bill. Those concerns revolved around the maintenance of the current coverage formula and bailout procedures, the twenty-five-year extension period, the new retrogression test, and what some Republican Senators considered the rushed process of deliberation that rejected any substantive amendments to the bill. (33) At various points it appeared that the legislation might be held over to the next Congress, especially once the language assistance provisions of section 203 became framed by the parallel debate over immigration reform. (34)
Any potential roadblocks to passage were removed once President Bush became involved, however. He scheduled his first presidential speech to the NAACP for July 20, 2006, and used the opportunity to stress his support for reauthorization "without amendment." (35) Majority Leader Bill Frist placed the House bill on the Senate calendar for the same day as the President's speech with rules that prevented any amendments on the floor. On the day before the Senate vote on the House bill, the Senate Judiciary Committee reported its own bill, which was identical to the House version, save for the addition of Cesar Chavez's name to the title. (36) This raised an interesting procedural question: if the Senate passed a bill that had a different title but exactly the same substance as the one passed by the House, would a conference committee nevertheless need to be assembled? To avoid that possibility, to prevent any Senate dillydallying on its bill, and to ensure the Senate vote would take place on the day of the President's speech to the NAACP, Majority Leader Frist simply moved the House bill to the Senate floor. (37) The Senate unanimously approved (98-0) the House bill shortly after the President's speech. (38)
Six days after the Senate passed the House bill and one day before the President signed the bill into law on July 27, the Judiciary Committee "Report" on its version of the bill was filed. (39) Not only was a presidential "stick" instrumental in propelling the House bill to passage in the Senate, but the unprecedented character of the after-the-fact committee report strongly suggests that the opportunity to alter the Senate Report provided a carrot that appeased some of the Senate Republicans who had reservations. In fact, despite a unanimous vote on the Senate bill both in the committee and on the floor, only half of the eighteen members of the Judiciary Committee- all of whom were Republicans-signed on to the Report. (40) The final draft of the Report itself was not circulated to Democratic senators on the Committee until the day the President signed the bill into law. (41) In their "additional views" included with the Report, the Democrats on the Committee emphasized, "We object and do not subscribe to this Committee Report ... which ... has become a very different document than the draft Report circulated by the Chairman on July 24, 2006." (42)
The evolution of the Senate Judiciary Committee Report offers the best window into the fragility of the political compromise that undergirds the new VRA and the basic disagreement that exists concerning its key provision. It also provides a unique case study in the self-conscious manipulation of legislative history for partisan ends and the shadow cast on the legislative bargaining process by the Supreme Court's recent federalism precedents. Moreover, given the importance the Court has attributed to legislative history in previous reauthorizations-namely, the centrality of the "Senate Factors" to the Court's subsequent interpretation of the 1982 Amendments to the VRA (43)--the unique character and procedural background of the Committee Report should cause concern regarding how courts or the Department of Justice (DOJ) might apply the law in concrete cases.
The hope of supporters of reauthorization was that the Senate Report would take the form of the House Report. Akin to a lawyer's brief, it would present the legislative record as unambiguously supporting reauthorization, and as providing substantial evidence to support its constitutionality. To do so, the Report would need to credit the damning examples of voting rights violations in covered jurisdictions and interpret the previous twenty-five years of experience under section 5 as demonstrating the VRA's continued utility. The proposed "Statement of Joint Views of Senate Judiciary Committee Members," which the Committee Democrats originally crafted but never released, did exactly that. (44) It is unsurprising that the Republican members of the Committee could not sign onto this "Statement of Joint Views." As is clear from the final product, the Republicans did not want to condemn the covered jurisdictions with as broad and resounding a declaration as did the House Committee. They also disagreed fundamentally with the Democrats' interpretation of the retrogression standard (45) and wanted to provide what they considered to be a more balanced view of the record, which would place greater emphasis on voting rights progress. The final Report bears no resemblance to the initial "Statement of Joint Views."
In their additional views in the final Report, the Democrats summarized their objections as follows: "we must register our disappointment that this Report does not reflect our views or those of scores of other cosponsors, does not properly describe the record supporting our bill, and does not fully endorse the bill we introduced and sponsored...." (46) The Democrats objected both to the body of the Report and to the additional views from Senators John Kyl, Tom Coburn, and John Cornyn. (47) The body of the Report (as well as the additional views of Senator Kyl) proffered a specific view of the new retrogression standard that protected only "naturally occurring majority-minority districts." (48) Many Democrats consider this interpretation of the central provision of the bill to be counter to their political interests, (49) if not subversive of the goal of protecting all kinds of districts (so-called influence or coalition districts) in which minorities usually constitute less than half of the district's voting population, but can nevertheless influence the outcome of an election. (50) There was a risk, expressed by some who testified at the hearings, that protecting the ability of minorities to elect their "preferred candidate of choice" would mean protecting their ability to elect Democrats, given that minorities tend to favor Democrats. (51) As described in greater detail below, the new retrogression standard can be seen as entrenching either Republican or Democratic gerrymanders depending on which types of districts it protects and which types of interdistrict population tradeoffs it prevents. The Republicans maintain that the law prevents minority population reductions in majority-minority districts and only in those districts. The Democrats would allow the unpacking of some such districts but would also protect certain minority-minority districts that elect minority-preferred candidates. (52)
In addition to their dispute over the meaning of the new section 5, the Democrats worried that the Report undermined the case supporting the constitutionality of the law. The inclusion of the additional views of Senators Cornyn and Coburn, in particular, had the potential to characterize the legislative record in a way that would lead the Supreme Court to strike down the law as exceeding Congress's power under the Enforcement Clauses of the Fourteenth and Fifteenth Amendments. (53) Those Senators expressed what several witnesses at the hearings found problematic with the Act: the lack of findings of relevant differences between the covered and noncovered jurisdictions, the alleged inflexibility of the bailout procedures, and the overreaching policy and constitutional precariousness of a twenty-five-year sunset period. (54) These concerns were heightened by what those Senators considered defects in the legislative process leading to reauthorization. Despite the fact that their party could have controlled the pace of the legislation (admittedly receiving criticism for being against civil rights had they slowed it down), they complained that "[a]n artificial rush to move the House version" occurred a full year before the relevant sections of the VRA were to expire. (55) "[T]his important legislation was-unfortunately-a bit of a foregone conclusion," they argued, and "[f]rom the outset the default seemed to be to accept the House product without deliberation." (56)
In contrast, the House Report had included an entire section arguing in favor of the constitutionality of the bill and rejected any specific clarifying definition of the new retrogression standard. (57) The House Report emphasized the long line of precedent upholding the VRA against constitutional challenge and compared the newly reauthorized VRA to the laws (such as the Family and Medical Leave Act (FMLA) (58) or equal access provisions of the Americans with Disability Act (59)) upheld after City of Boerne v. Flores. (60) Those laws were constitutional because they protected suspect classes or fundamental rights; the Voting Rights Act has the advantage of doing both. The House Report also evaded the tough questions concerning the retrogression standard by merely clarifying (and reiterating ad nauseum) that it overruled Georgia v. Ashcroft, (61) reinstated the standard from Beer v. United States, (62) and focused the retrogression inquiry on the "ability to elect" rather than on any amorphous standard of influence. It did not give guidance as to how the ability-to-elect determination should be made. (63)
Fundamental disagreements existed among the senators over the desirability, constitutionality, and the meaning of the law. Thus, even an attempt to pass a watered-down consensus report and to leave heated disagreements for expressions of "additional views" was not possible. As is clear from the final Report, some Republican senators found the geographic reach of the law to be unfair and potentially unconstitutional, as well as unjustified given what they regarded as voting rights progress. On the fundamental question of what the major new requirement in the law (the retrogression standard) meant, the Republicans believed it only protected "naturally occurring majority-minority districts," (64) while the Democrats considered it to protect a greater variety of districts with varying percentages of racial minorities. (65) The parties could not even agree as to what evidence of voting rights violations should be included in the body of the Report, as opposed to the Appendix.
These substantive disagreements and procedural irregularities may lead reviewing courts to disregard the Senate Report entirely, and rely on the House Committee Report and record, the findings in the law itself, and the materials that the senators had before them at the time of their vote. (66) Indeed, the gamesmanship surrounding the Committee Report will provide additional ammunition for those, such as Justice Scalia, who never want to consider such legislative materials as a guide to statutory meaning. (67) However, as discussed later in this Article, on certain key questions of what the law means, the Senate Report provides the principal legislative history as to how the law should work in practice. Therefore, while judges may discount it in their opinions, they inevitably will need to confront it.
II. THE EVIDENTIARY RECORD
The differing views of policymakers and advocates as to what the law actually meant remained pushed to the background of the legislative debate because Congress spent most of its time and effort assembling a record sufficient to justify the constitutionality of the law. The constitutional standard for congressional authority to enforce civil rights had changed since the last reauthorization. Supporters of the Act were justifiably concerned that the new VRA (whatever it provided) was more vulnerable than its predecessors.
It is worth mentioning at the outset, however, the unique constitutional quandary presented to supporters of the Act concerning the record they needed to develop. (68) The new constitutional standard for congressional authority to enforce civil rights that emerged from City of Boerne v. Flores (69) and its progeny (70) established that such laws must be "congruent and proportional" to remedying or preventing violations of the Fourteenth Amendment. The new VRA was quite different from other laws either upheld (71) or struck down (72) post-Boerne: (1) the bill proposed renewal of existing legislation, not drafting a law from scratch; and (2) the law would not apply nationwide. (73)
First, because Congress was not writing legislation anew, but rather reauthorizing a law currently in effect, it was unclear what record of discrimination it needed to develop. (74) To prove the law was necessary, the best evidence would be data concerning the extent of voting rights violations in the covered jurisdictions, especially if such violations were more prevalent in covered than in noncovered jurisdictions. However, if the Act was working well, then few such examples should exist. Conversely, if widespread voting rights violations continued in the covered jurisdictions, then the law arguably was not working, and it would be difficult to justify it as a congruent and proportional remedy. (75) Like those who advocate for continuing a heightened police presence in a previously, but not presently, high-crime neighborhood, advocates for renewal needed to marshal evidence both of the law's success and the harmful consequences of its removal. (76)
Second, "congruence and proportionality" may mean something different for a law that does not apply to the nation as a whole. Unlike a law of general applicability, a law with a specific geographic reach may need to justify not only the need for the law but also the differential need for the law in some areas rather than others. Does the constitutional test require a record merely of continued voting rights violations in the covered jurisdictions? Or does it require a record of a greater frequency, tendency, or severity of such violations in the covered, as opposed to uncovered, jurisdictions? (77) If the latter, then the concerns expressed above become exacerbated, as an absence of differences between the two classes of jurisdictions could either constitute evidence of the law's successful deterrent effect or it might suggest the arbitrariness of the geographic choice the law makes. (78)
Supporters of reauthorization decided that the safest course of action was to stick with the coverage formula that the Supreme Court had previously upheld. Despite the recognized need to extend coverage to the newest generation of voting rights violators, constitutional and political constraints prevented any alteration of the statute's geographic reach. The requirements of the Boerne standard were sufficiently uncertain that any change in the coverage formula was seen as an additional gamble on the ability of Congress to predict what types of evidence the Court would find important. Moreover, the political hurdles of adding new jurisdictions, most of which likely would have been Republican, to the preclearance regime made maintaining current coverage the safest political choice. The new Act was therefore not an attempt to capture all of the worst voting rights violators, but rather an effort to capture some of them and to preserve historic gains where they had been made. As a result, supporters of the Act sought to develop an evidentiary record for the principal purpose of explaining why the covered jurisdictions should remain covered, rather than justifying the coverage of certain jurisdictions but not others.
Despite considerable disagreement as to the sufficiency of the legislative record for purposes of proving the law's constitutionality, very little disagreement existed as to the facts on the ground. The legislative record assembled principally in the House, (79) but repeated in the Senate, contains four categories of evidence to support reauthorization of the VRA: (1) statistics as to minority voter turnout, registration, and rates of officeholding; (2) statistics concerning DOJ and jurisdiction behavior with respect to the preclearance process; (3) examples of voting rights violations in the covered jurisdictions; and (4) data as to all VRA section 2 litigation nationwide. The hearings made clear that the supporters and opponents of the bill differed principally as to how to interpret and emphasize the evidence.
A. Rates of Minority Voter Turnout, Registration, and Officeholding
For the original VRA and the subsequent two reauthorizations, voter turnout statistics represented one of two critical components to deciding which jurisdictions should be covered. A jurisdiction with a voter turnout rate below 50% that also employed a "test or device," such as a literacy test, in the 1964 election would be covered under the original section 5. (80) This "trigger" or coverage formula was reverse engineered in order to capture a foreseeable group of jurisdictions, principally in the South. (81) This original trigger, as well as a subsequent amendment that included in the definition of "test or device" English-only ballot materials in jurisdictions with a large number of non-English speakers, was both overinclusive and underinclusive of discriminatory jurisdictions. (82) However, the formula roughly corresponded to jurisdictions Congress found "guilty" and avoided the politically fraught task of merely picking and choosing jurisdictions based on subjective judgments about their relative lack of protection for minority voting rights. (83) Moreover, any jurisdictions erroneously captured by the coverage formula could simply "bail out" by demonstrating in court their unproblematic voting rights records.
For the 2006 reauthorization, voter registration and turnout statistics--either aggregate registration and turnout by state or differential registration and turnout by racial group-did not help the cause much. (84) Turnout rates in the covered and uncovered jurisdictions do not differ consistently. (85) Both the House and the Senate Reports also noted the remarkable decrease in differential registration and turnout rates among racial groups. The Senate Report emphasized success, noting that in some covered jurisdictions (California, Georgia, North Carolina, Mississippi, and Texas) African American turnout exceeded that of whites. (86) The House Report also noted success, but made express findings as to differential turnout rates for African Americans in Virginia and South Carolina, and Hispanics in Texas and Florida, that provided evidence of the continued need for section 5. (87)
Remarkably, both the House and Senate reports, when comparing black and white turnout, count Hispanics as white instead of comparing non-Hispanic white and black turnout. As a result, white turnout appears artificially low and in some states makes black turnout appear to be a few percentage points higher than white turnout. For example, the Senate Report notes higher black turnout than white turnout in 2004 in Texas: blacks (55.8%), whites (50.6%), Hispanics (29.3%). (88) Those statistics are technically correct, yet once Hispanics are taken out of the white category the picture changes considerably. Non-Hispanic-white turnout in Texas in 2004 was 63.4% (89)-7.6 percentage points higher than black turnout. (90)
Although one can focus on a few bad states or aberrant years, turnout statistics no longer capture the level of unconstitutional discrimination that may exist in the covered or noncovered jurisdictions. African Americans generally trail whites in turnout by between five and ten percentage points. (91) States that disenfranchise felons prevent a substantial share of the African American male population from voting, but the practice does not appear to affect differential turnout between the covered and uncovered jurisdictions. (92) Hispanics' extremely low voter turnout rates are due in large part to their ineligibility to vote given their low rates of citizenship. (93) The voter turnout statistics presented in the House and Senate reports use voting age population as the denominator, as opposed to citizen voting age population or eligible voters. Because a large number of Hispanics in the voting population are ineligible to register even if they wanted to, their turnout rates appear dramatically lower than other racial groups.
Data in the House and Senate reports with respect to minority officeholding reflected a similar pattern. Although minorities hold only about 5% of elected positions nationwide, (94) as the committee reports noted, there has been a very significant increase in the number of African Americans and Hispanics holding positions in Congress and state legislatures. (95) The House Report, however, noted minorities' underrepresentation as compared to their share of the population. It emphasized that in the Deep South (Alabama, Georgia, Louisiana, Mississippi, South Carolina, and North Carolina), where African Americans make up 35% of the population, they hold only 20.7% of the seats in states legislatures-with even less frequent success in winning statewide office. (96) The Senate Report struck a somewhat different tone, pointing to near proportional representation for African Americans in the state legislatures and congressional delegations from Georgia and Mississippi. (97) Nothing in the record, however, pointed to a difference in rates of minority officeholding between covered and uncovered jurisdictions. (98)
B. Evidence Concerning the Preclearance Process: Rates of Submissions, Denials, and Requests for More Information
The threat to reauthorization posed by the success of the VRA becomes clearest when considering the evidence concerning violations of section 5 itself. (99) Preclearance behavior is the most easily measured and presented data in the record, which contains rates of preclearance submissions, denials, and requests for more information, as well as accounts as to the deterrent effect of section 5.
Despite large increases in the volume of preclearance requests, the rate and absolute number of DOJ denials of preclearance have declined in recent years. Although the total number of preclearance denials (682) was greater for the twenty-five years after the 1982 amendments than during the first seventeen years of the VRA, (100) the rate of DOJ objections to preclearance requests has decreased from over 4% in the first five years after the Voting Rights Act, to between 0.05% and 0.23% from 1983 to 2002. (101) With only ninety-two total objections in the last ten years, the annual objection rate since the mid-1990s has dropped to an average of less than 0.2%. (102)
Much of that recent drop is due to changes in DOJ enforcement of the discriminatory purpose prong of section 5 as mandated by the Supreme Court in Reno v. Bossier Parish (Bossier Parish II), (103) which the VRARA overturns. Nearly three-quarters of the objections from 2990 to 2000 involved purposeful discrimination, with 43% based on purpose alone. (104) However, the Bossier Parish II decision rejected discriminatory purpose (as opposed to retrogressive purpose) as the purpose inquiry for section 5. Since that decision, the DOJ has lodged only a handful of purpose-based objections. (105)
A better indicator of section 5's deterrent effect, however, might be the number of DOJ "Requests for More Information" (MIRs) and the rate of withdrawal of voting changes pursuant to such requests. (106) After all, the fact that the preclearance regime leads to few denials could simply mean that the section is working as intended. Like any other law, section 5's effectiveness should not be evaluated by the number of times it is broken. The Voting Section of the Civil Rights Division files a "Request for More Information" when the submission from the jurisdiction does not provide all the information needed to evaluate the potential retrogressive effect of a voting change. (107) An MIR also represents a DOJ signal that the voting change might be found retrogressive (and denied preclearance) unless the jurisdiction allays the DOJ's concerns. Since 1982, DOJ has sent over 800 requests for more information regarding voting changes, leading jurisdictions to withdraw their submissions in 205 instances and change their submissions in many others. (108) Again, as a total share of preclearance submissions, this represents a small fraction, but it gives a sense of how many dogs did not bark as a result of the threat of a denial of preclearance.
The same could be said regarding the testimony pointing to DOJ negotiations with jurisdictions even before they submit a plan for preclearance. Often a jurisdiction will work with the DOJ to ensure that a voting change that might be retrogressive is altered, even before it is submitted, in a way that ensures a grant of preclearance. These bargains occur in the shadow of section 5 without the threat of a preclearance denial needing to be exercised. (109)
C. Voting Rights Violations in Covered Jurisdictions
The hearings in the House and the Senate, as well as the committee reports, are replete with examples...
|
